1. Field of the Invention
This invention relates generally to a battery charger and a method of charging a rechargeable battery, specifically alkaline secondary batteries, Ni--Cd (Nickel-Cadmium), Ni--H.sub.2 (Nickel-Hydrogen) and Ni--MeH (Nickel-Metal Hydride) batteries which utilize an inflection in chemical polarization voltage as a signal to terminate charging.
2. Background and Description of the Related Art
A large number of portable electronic devices, e.g. cellular phones, two-way radios, laptop computers and camcorders, are in widespread use. In addition, portable electronic devices have become more common in the aerospace industry. With growing use of these devices, it has become necessary to provide a rechargeable battery that is in a continuous state of operational readiness. In this regard, it is preferable to utilize a rechargeable battery that uses a method of recharging that avoids both under- and overcharging.
Overcharge conditions are undesirable and disadvantageous to cycle life of the battery. Overcharge reactions in vented cells result in electrolysis and loss of water that must be replaced; in sealed cells, overcharge reactions create pressure and heat since the recombination reactions of gases produced during overcharge are exothermic.
Overcharge reactions that result from high charging voltage can be prevented by simply limiting the charging voltage to a certain value. This simple approach is, unfortunately, only partially successful with certain cell types, e.g., lead-acid cells, vented Nickel-Cadmium cells, and sealed lithium ion cells.
Despite the adverse consequences of overcharging, in conventional charging methods, overcharge reactions are tolerated to complete the charging process. Sealed cells capable of recombining the overcharge reaction products are able to tolerate overcharge at low rates because the pressures of by-product gases are low and the heat generation is slow enough that the generated heat can be easily dissipated and lost. If no electrolyte constituent is lost, it is less crucial to terminate charging. However, continuous overcharge, even at low rates, reduces the life cycle of the cells.
Rapid charging, i.e., charging in less than one hour, presents much more of a challenge in both vented and sealed cells. The first problem results from the limited rate of charge distribution or equilibration within the electrode plates. Also, at higher charge rates, the overcharge reactions begin to appear at a lower fraction of full charge. When overcharge reactions appear, the efficiency of the charge reactions declines and much of the columbic energy is wasted on the overcharge reactions. The rapid heating of a battery during high-rate overcharge may cause safe pressure limits to be exceeded, as well as cell venting.
U.S. Pat. No. 5,352,967, which is hereby incorporated herein by reference in its entirety, provides a good summary of some of the known methods of charging storage batteries. Most methods of charging sealed Ni-Cad and NiMH batteries use voltage and temperatures as parameters for terminating the charging phase. The methods disclosed therein focus on various techniques of determining proper charge termination and include: constant current mode; delta temperature/delta time mode (dT/dt); negative delta voltage charge mode (-.DELTA.V); positive delta voltage or delta voltage/delta time mode (dV/dt); pulse charge mode; and reflex mode.
In the constant current charge mode, the battery is continuously overcharged with a low current. Although the expenditure for a constant current source is relatively low, the long charging time causes damage to the cell. In the constant charge mode, it is customary to restrict the charging time. Therefore, as soon as a selected time has elapsed, the charging operation is terminated. Accordingly, since the constant current charge mode does not take into account the charge condition or chemical makeup of the cell, under- or overcharging of the cell can result.
In delta temperature/delta time charging (dT/dt), the charging current is switched off once the cell has reached a predetermined slope in the temperature versus time curve for the cell This method can generate incorrect termination signals. The charging process may be terminated prior to the battery being fully charged if the preset value of dT/dt is too low, or, conversely, the charging process may be terminated too late if the value of dT/dt is too large.
In the negative delta voltage charge mode, a negative slope in the charge curve (dV/dt&lt;0), which appears after complete charging of the battery, is used to terminate charging. The batteries are charged from a constant current source, and the charging voltage rises steadily for as long as the cells are capable of converting the supplied energy into chemical energy. When the cells are no longer capable of storing the supplied energy, the supplied energy is converted into heat, and the cell voltage drops. The drop in the cell voltage is used as an indicator to terminate charge. However, fluctuations in the battery voltage caused by surface effects may be erroneously interpreted as a signal to terminate charge. Therefore, premature break-off of the charging operation is often seen when utilizing the negative delta voltage charge mode. Another problem with negative delta voltage charge mode is that with low rates of charge, the battery does not experience a negative delta voltage in the end of charge. Additionally, Nickel-Metal Hydride batteries do not have the pronounced charging voltage curves seen in Ni--Cd batteries, and as a result, are often overcharged using the negative delta voltage charge mode.
In the positive delta voltage or delta voltage/delta time mode (dV/dt), the slope of the voltage charging curve is evaluated to determine when to terminate charging of the battery. Theoretically, the rise in the charging voltage decreases when the battery is near full charge. Utilizing a mathematical differential of the charge curve, the reduction in this rise in the charging voltage can be evaluated as the criteria for terminating charging. This method suffers from the fact that the difference in the rise may not be dramatic enough to cause termination of the charging at a proper time with ensuing overcharge of the battery. Additionally, due to fluctuation in the charging curve, this first order derivative charging method may terminate charging prematurely.
The methods described above are based, at least in part, upon the phenomena associated with oxygen production on the nickel oxide electrodes and oxygen consumption at the cadmium electrode (or complementary electrode). Oxygen production causes voltage to peak and creates inflection points. Oxygen consumption is responsible for heat generation and the decrease of internal impedance, resulting in a decrease in charging voltage. The accompanying rise in temperature can also be used for charge termination. One of the problems with relying on oxygen production as an indicator of charge termination is that oxygen production causes voltage peak and inflection points. To overcome this problem, it has been described in the related art to use certain open circuit voltage parameters as indicators of charge termination.
Pulse charging utilizes a high current charge followed by an interruption period. The interruption period allows the voltage of the battery to be determined during a currentless phase or under open circuit voltage conditions (OCV) in order to determine the open circuit voltage of the battery (V.sub.OCV).
U.S. Pat. No. 5,477,125, which is hereby incorporated herein by reference in its entirety, proposes a method of recharging a battery that comprises the steps of periodically interrupting the charge current, sampling the resistance free voltage (V.sub.o) after the delay period, and determining the point or points on the V.sub.o (t), dV.sub.o /dt and d.sup.2 V.sub.o /dt.sup.2 curves to detect points indicative of charge termination. The current measuring periods are repeated every 10 seconds, and a delay period after current interruption is used to allow the battery to reach an equilibrium. This reference utilizes a pre-set VREF and this may result in a large degree of error due to the fact that the battery rate of charge and design criteria are not generally accounted for when selecting V.sub.REF.
Each of the above-described open circuit monitoring techniques suffer from the disadvantage that they do not precisely indicate the point or points necessary for charge termination. To prevent overcharge more accurately, it is necessary to measure and detect a relatively small change in open circuit voltage (i.e., 30-40 mV) and detect a relatively large change in absolute charging voltage (i.e., 8-9V). Use of the above described charging schemes normally does not recognize such a small change in open circuit voltage. Often, with the charging techniques of the related art, such a small change cannot be recognized above the "noise" or artifacts of the charging process.